Mexican Americans (MAs) suffer more from stroke than non-Hispanic whites (NHWs). Ischemic stroke is more common in MAs and their neurologic, functional and cognitive outcomes after stroke are worse than in NHWs. The reasons for the disparity in post-stroke outcome are unclear. Pre-stroke function and initial stroke severity are similar between the two groups as are ischemic stroke sub-types. One potential explanation for the worse post-stroke neurologic, functional and cognitive outcome in MAs compared with NHWs is allocation and effectiveness of post-stroke rehabilitation. There is remarkably little data demonstrating whether rehabilitation is dosed differently for MAs compared with NHWs, and still less information about whether, for a given dose of rehabilitative services following stroke, there is differential benefit by ethnicity. The current application will utilize the existing population-based Brain Attack Surveillance in Corpus Christi (BASIC, NSR0138916) project's infrastructure and strong community relations to develop and pilot a method to collect the necessary data to determine the role of rehabilitation in ethnic disparities in post-stroke outcomes. Previous studies have suggested that looking at overall time spent in rehabilitation does not predict post-stroke outcome. However, specific components of physical, occupational and speech therapy, a practice-based approach, has been shown to be associated with stroke outcomes, and these associations have been shown to vary by race. However, this practice-based approach has not been implemented in a population-based manner across the range of settings where stroke patients receive rehabilitation services, and no study has used this approach in an ethnically diverse population. Therefore, our plan is to build on previous work by developing and utilizing a practice-based design in our population-based stroke study. Specifically, we will 1) continue to build the needed relationships with rehabilitation service providers in the community; 2) work with local rehabilitation therapists to refine data collection instruments as part of the practice-based design; 3) pilot test data collection of specific rehabilitation components of post-stroke rehabilitation across all rehabilitation settings; and 4) analyze this data to determine the feasibility of this approach for a larger study and to provide preliminary data on differences in access and effectiveness by ethnicity. In total, our infrastructure development, refinement of tools to measure specific therapy modalities and pilot testing will position us perfectly to submit an R01 application to identify ethnic differences in access to rehabilitation and specific rehabilitation services associated with improved functional outcome in MAs and NHWs.